| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
Interact CardioVasc Thorac Surg 2005;4:256-259. doi:10.1510/icvts.2004.101006 © 2005 European Association of Cardio-Thoracic Surgery
Classification for predicting mediastinal lymph node metastases in patients with T1 or T2 lung cancerDepartment of General Thoracic Surgery, Hôpital Universitaire, 2 bd Marechal de Lattre de tassigny, BP 1542, 21034 Dijon Cedex, France Received 22 October 2004; received in revised form 17 February 2005; accepted 21 February 2005
*Corresponding author. Tel.: +33 3 80 29 37 47; fax: +33 3 80 29 35 91.
The aim of this study is to classify patients into risk groups for mediastinal lymph node metastases. Three hundred and thirty-seven patients underwent lung resection for lung cancer. The nodal status was pN0 in 181 patients, pN1 in 62 and pN2 in 94. The presence of the involvement of one mediastinal compartment (superior or inferior) or two mediastinal compartments (superior and inferior) was considered to be the main end point. One mediastinal compartment was involved in 65 patients and two mediastinal compartments in 29 patients. Two variables (visceral pleural invasion and the primary tumor location) were retained in the model. The regression tree analysis categorized patients into 3 risk groups for the involvement of two mediastinal compartments. The low-risk group included 118 patients with a tumor located in the left side and no visceral pleural invasion. The intermediate-risk group included 160 patients with a tumor located in the right side and no visceral pleural invasion. The high-risk group included 59 patients with visceral pleural invasion and a tumor located in the right side or left lower lobe. A practical, easy-to-use risk grouping system is proposed to aid the decision making and to simplify mediastinal lymphadenectomy procedure.
Key Words: Lung cancer; Mediastinal lymph node metastases; Lymphadenectomy; Algorithm
Surgical resection is the treatment of choice for patients with T1 or T2 lung cancer. However, the extent of mediastinal lymphadenectomy in the treatment of non-small-cell lung cancer is still a matter of controversy. While some centers perform mediastinal lymph node sampling with resection of only suspicious lymph nodes, others recommend systematic radical mediastinal lymphadenectomy to improve survival and achieve better staging. Radical mediastinal lymphadenectomy enables the clinician to determine the extent of invasion more precisely and often reveals pN2 at multiple levels [1]. The therapeutic value of radical lymphadenectomy remains a subject of debate [2]. Radical lymphadenectomy makes it possible to identify individual lymph node stations according to the location of the primary tumor. However, there have been few reports regarding mediastinal lymph node metastases according to the location of the primary tumor. Indeed, better knowledge of mediastinal lymph nodes metastases could avoid systematic dissection of all of the mediastinal compartments. This article analyses the appearance of metastases in mediastinal lymph nodes with regard to the site of the primary tumor and their correlation with the clinicopathologic properties of the primary tumor. The aim of this study is to classify patients into risk groups for mediastinal lymph node metastases.
From January 1997 through December 2002, 337 patients underwent pulmonary resection for T1 or T2 lung cancer: lobectomy in 254 patients, bilobectomy in 16 patients and pneumonectomy in 67. Patients were excluded from this study according to the following criteria: (1) limited resection; (2) exploratory thoracotomy; (3) mediastinal lymph nodes with a short axis diameter of more than 10 mm measured on CT scan; and (4) preoperative chemotherapy. The characteristics of the patients are reported in Table 1.
Lung resection was systematically combined with radical en-bloc mediastinal lymphadenectomy as described by Martini [3]. Briefly, the superior mediastinal compartment contained between the trachea, the superior vena cava from the level of the azygos vein to the right subclavian artery, and the right recurrent laryngeal nerve was dissected and the trachea, azygos vein, superior vena cava, and ascending aorta were completeley freed from all tissue. The subcarinal, paraoesophageal, and inferior pulmonary lymph nodes were removed en bloc, exposing the entire thoracic oesophagus and the vagal nerve. In left-sided cancer, the subaortic compartment contained between the left pulmonary artery, the aortic arch, the left recurrent laryngeal nerve, and the phrenic nerve was dissected by completely freeing the left vagal nerve and the recurrent laryngeal nerve. 2.1. Outcomes Determining the number of mediastinal lymph node compartments involved was considered to be the main end point. Lymph nodes of region 2, 3, 4, 5 and 6 were classified in the superior mediastinal compartment, whereas lymph nodes of region 7, 8 and 9 in the inferior mediastinal compartment. This main end point included 3 categories: (1) N0 or N1, (2) Involvement of one mediastinal lymph node compartment (superior or inferior mediastinal compartment), and (3) Involvement of two mediastinal lymph nodes compartment (superior and inferior mediastinal compartment).
Prognostic factors for mediastinal lymph node involvement were identified through univariate analysis using
A total of 4559 lymph nodes were removed (average: 13 per patient). A total of 2997 mediastinal lymph nodes were removed (average: 9 per patient). The distribution of mediastinal lymph node involvement according to the location of the primary tumor is reported in Table 2. Metastases in both superior and inferior mediastinal lymph node compartments were most frequently to tumors located in the right side (Table 2).
One mediastinal compartment was involved in 65 patients (19%) and two mediastinal compartments in 29 patients (9%). The clinicopathologic variables associated with mediastinal compartment involvement in univariate analysis are reported in Table 3. In multivariate analysis, visceral pleural invasion and primary tumor location remained significant.
Next, the regression tree analysis categorized patients into 3 risk groups for two mediastinal compartment involvement. The Low-risk group included 118 patients with tumors located in the left side and no visceral pleural invasion. The Intermediate-risk group included 160 patients with tumors located in the right side and no visceral pleural invasion. This risk group also included patients with tumors located in the left upper lobe and visceral pleural invasion. The High-risk group included 59 patients with tumors located in the right side or in left lower lobe and visceral pleural invasion. If the tumor was located in the left side without visceral pleural invasion, the risk of lymph node metastases in both superior and inferior mediastinal compartments was only 3% (Fig. 1). For patients with tumors located in the right side without visceral pleural invasion the incidence of two mediastinal compartment involvement was 8% (Fig. 1). Finally, when the visceral pleura were invaded by the tumor, the incidence of lymph node metastases in both superior and inferior mediastinal compartment was evaluated at 20% (Fig. 1).
The number of lymph nodes removed after a mediastina lymphadenectomy varied from patient to patient. However, the average number of lymph nodes removed between 7 and 9 per case [1,6]. Nodal metastases in the upper mediastinum were frequently observed as were metastases in the lower mediastinum [7,8]. Metastases in the mediastinal lymph node were significantly more frequent on the right side than on the left. The risk of the involvement of two mediastinal compartments was tripled when the tumor was located in the right side. In our study, for the tumor located in the upper lobe, lymph node metastases were more frequently sited in the superior mediastinal compartment, and for the tumor located in the lower lobe in the inferior mediastinal compartment. The invasion of visceral pleural significantly increased the risk of lymph node involvement [9]. The Grade of differenciation with regard to risk of mediastinal lymph node involvement is reported in the literature [9]. In our study, this factor was not studied because it is not a clinical variable. During the operation, the surgeon can determine if the tumor has invaded the visceral pleural. Factors included in our model are clinical and easily usable by the surgeon during the procedure to make the decision. Radical lymphadenectomy at the time of lung resection is essential to achieve accurate staging [1,2,6]. However, the therapeutic value of radical lymphadenectomy is uncertain. Only one study [6] showed an improvement in the survival after radical lymphadenectomy. This prospective randomised trial [6] did not describe the technique of radical lymphadenectomy and the procedure used in arm control. Radical block dissection of all mediastinal lymphatic vessels increases the duration of the operation and this procedure could be responsible for postoperative complications such as bronchopleural fistula. This type of classification was used for other cancers including that of the prostate [10]. Our model is easy to use because two clinical variables, the location of the tumor and visceral pleural invasion were included. For a patient with a tumor located in the left side without visceral pleural invasion, it is useless to perform systematic lymphadenectomy of both superior and inferior mediastinal lymph node compartments. This combination accounted for almost 35% of 377 patients into the low-risk group, only 3% were badly classified. In contrast, when the tumor is located in the right side with visceral pleural invasion, radical lymphadenectomy has to be performed in both superior and lower mediastinal compartments. Decisions concerning the intermediate-risk group are more difficult because in 8% of cases two mediastinal compartments are invaded. The rationale of this model is based on the involvement of mediastinal lymph nodes according to the characteristics of the lung tumor. Radical mediastinal lymphadenectomy in all patients with lung cancer is perhaps unnecessary. This procedure increases duration of the operation. Radical block dissection of all mediastinal lymphatic vessels might occasionally damage the right or left recurrent laryngeal nerve [3]. Chylothorax is possible after lymph node dissection [3]. Bleeding is generally minimal. Radical mediastinal lymphadenectomy in the right side could be considered to be a risk factor of ARDS [11]. The development of this type of model to simplify the procedure and reduce postoperative morbidity has already been confirmed by other studies for prostate carcinoma [10]. However, it is necessary to carry out a validation study with an independent sample to assess the performance of this model. In conclusion, lung cancer without visceral pleural invasion does not justify a systematic lymphadenectomy of both superior and inferior mediastinal lymph node compartments. Finally, this model should be validated by another study.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
Top
Abstract
1. Introduction
2 tests. After univariate analysis, the variables were included in multivariate logistic regression for multinomial data to identify prognostic factors 